Adaptive NOMA MAC Architecture for Joint Modulation Scheduling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing wireless communication systems lack a medium access control (MAC) architecture for deploying non-orthogonal multiple access (NOMA) techniques, which are beneficial in scenarios where multiple devices share resources under a single entity's control, in mobile edge computing, and in device-to-device communications without massive antenna arrays.
Innovation Solution
Development of MAC architectures that utilize adaptive NOMA modulation, including end-to-end and codebook-based approaches, which involve a MAC scheduler, parameter estimation, device-specific link adaptation, and modulation and coding blocks, leveraging machine learning for optimal modulation and demodulation schemes.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If NOMA techniques are deployed without a dedicated MAC architecture, then device complexity is reduced, but communication rates and resource utilization deteriorate
Solution Approach 1:
The MAC architecture is segmented into distinct functional blocks: a MAC scheduler for resource allocation, a parameter estimation block for channel state information, a device-specific link adaptation block for individual device optimization, and a joint adaptation block for coordinated multi-device transmission. This segmentation enables complex NOMA operations while maintaining modular design that manages complexity.
Solution Approach 2:
The MAC architecture implements dynamic adaptation through machine learning models that continuously adjust modulation schemes, coding rates, and power allocation based on real-time channel conditions. The link adaptation blocks dynamically optimize transmission parameters for each device and for joint multi-device transmissions, enabling high communication rates under varying network conditions.
2Productivity
If adaptive NOMA modulation is implemented, then resource utilization improves, but device complexity increases
Solution Approach 1:
The MAC scheduler performs preliminary resource allocation and power distribution before transmission, estimating channel parameters in advance. The device-specific link adaptation blocks pre-compute optimal modulation and coding schemes based on estimated channel conditions, reducing real-time computational complexity while maximizing resource utilization.
Solution Approach 2:
Machine learning models serve as intermediaries between channel conditions and transmission parameter selection. The models translate complex channel state information into optimized modulation schemes and coding rates, simplifying the adaptation process while achieving high resource utilization through intelligent parameter selection.
3Productivity
If machine learning is used for modulation scheme selection, then communication rates enhance, but processing time increases
Solution Approach 1:
Machine learning models are trained offline on extensive channel condition datasets to learn optimal modulation scheme selections. During runtime, the pre-trained models perform rapid inference based on current channel estimates, avoiding time-consuming real-time optimization calculations while maintaining high communication rates through intelligent scheme selection.
Data Source
AI summary
A device includes a wireless transceiver and a processor. The processor is configured to determine a direction of transmission for a non-orthogonal multiple access (NOMA) transmission; determine a set of resources to be used for the NOMA transmission; determine a forward error correction (FEC) coding scheme to be used for the NOMA transmission; determine, using a codebook, a joint modulation scheme to be used for the NOMA transmission; and transmit or receive the NOMA transmission on the set of resources, in accord with the direction of transmission, the FEC coding scheme, and the joint modulation scheme.